Short Notes

Research Methods (STA 630)

Lesson No.1
Introduction, Definition, & Value of Research
Q what is research and whats the vale o! research"
Soltion
Research is something to do with the laboratory where scientists are apparently doing
some experiments / the process of finding solutions to a problem after through studying and
analyzing the situational factors. It is gathering information needed to answer a uestion, and
thereby help in sol!ing a problem.
#ale o! Research$
"he nature of research problems could !ary. #roblems may refer to some undesirable situation
or these may refer to simply a snooping of the research that may be agitating his or her mind.
$or example, in a recent %&/%' examination of the #un(ab )ni!ersity *+ percent of the students
failed. "hat is a colossal wastage of the resources, hence an undesirable situation that needs
research to find a solution. "he researchers may come up with a !ariety of reasons that may
relate with the students, the teachers, the curricula, the a!ailability of boo,s, the examination
system, the family en!ironment of the student, and many more. 'o a study may be carried out
diagnose the situation, and the recommendations to be applied to o!ercome the undesirable
situation of mass failure of students.
"he !alue of research for policy ma,ers, planners, business managers, and other sta,eholders
is that it reduces uncertainty by pro!iding information that impro!es the decision ma,ing
process. Decision ma,ing process associated with the de!elopment and implementation of a
strategy in!ol!es four interrelated stages,
Identifying problems or opportunities
Diagnosing and accessing problems or opportunities
'electing and implementing a course of action
-!aluating the course of action
Lesson No.%
'cientific .ethod of Research & Its 'pecial $eatures
&hat is Scienti!ic 'ethod and write down its !eatres or characteristics"
Soltion
'cience is a way to produce ,nowledge, which is based on truth and attempts to be uni!ersal.
'cience is a method or procedure to produce ,nowledge, which could be used for the solution
of problems as well as for the generation of uni!ersal theories, principles and laws through the
process of obser!ation and re obser!ation. /bser!ation here means that scientists use 0sensory
experiences1 for the study of phenomenon. "hey not only do the obser!ation of a phenomenon
but also repeat the obser!ation se!eral time because they want to definite and positi!e about
their findings, rather the results are organized, systematized, and made part of the existing body
of ,nowledge, all this procedure for the creation of ,nowledge is called 0scientific method1.
(')ortant *eatres or +haracteristics o! Scienti!ic Method
,')irical2 'cientific method is concerned with the realities that are obser!able through
sensory experience and generates ,nowledge which !erifiable by experience or obser!ation.
'ome of the realities could be obser!ed directly but some are not obser!ed directly but
researchers ha!e designed ways to obser!e these indirectly.
#eri!ia-le2 3nowledge obser!es through scientific method again !erified by the same
researcher or other using senses to confirm, and place more faith and confidence in those
findings or conclusions, If similar findings emerge on the basis of data collected by other
researchers using the same method. "hey will gain confidence in the scientific nature of our
research.
+'lative2 prior to the start of any study the researchers try to scan through literature and
see that their study is not a repetition in ignorance. Instead of rein!esting the wheel the
researchers ta,e stoc, of the existing body of ,nowledge and create a lin,age between the
present and pre!ious body of ,nowledge has to be established and that is how the ,nowledge
accumulates.
.eter'inistic2 'cience is based on the assumption that all e!ents ha!e antecedent causes
that are sub(ect to identification and logical understanding. "he scientific researchers try to
explain the emerging phenomenon by identifying its reasons. "he researcher tries to narrow
down the large number of reasons which implies the explanation with the minimum number of
!ariables that are responsible for an undesirable situation in such a way some action could
ta,en.
,thical and (deolo/ical 0-1ecti!ity2 "he conclusions drawn though interpretation of the results
of data analysis should be ob(ecti!e that is they should be based on the facts of the finding
deri!ed from actual data and not on our emotional !alues. &ny interference of their personal
li,ings and dis li,ing in their research can contaminate the purity of data, which ultimately can
affect the predictions made by the researcher.
Statistical 2enerali3ation2 4eneralizability refers to the scope of the research findings in one
organization setting to other settings. $or wider 4eneralizability, the researcher sampling design
has to be logically de!eloped and a number of other details in the data collection methods need
to be carefully followed. 5ere the use of statistics becomes !ery helpful in ma,ing
generalizations which is one of the goals of scientific method because it is a de!ice for
comparing what is obser!ed and what is logically expected.
,4)lanation2 'cience is fundamentally a explanation acti!ity and scientific explanation must
ma,e sense.
(ndctive Lo/ic2 /ne starts from obser!ed data and de!elops a generalization which explains
the relationships between the ob(ects obser!ed.
.edctive Lo/ic2 /ne starts from some general law and applies it to a particular instance.
Important features of scientific method say that there are two )ower -ases o! scienti!ic
5nowled/e 6i7 -mpiricism such as 'ensory -xperience or /bser!ation and 6ii7 Rationalism such
as the logic explanations for regularity and then conseuence ional argumentation for ma,ing
generalizations.
,4)eri'ental .esi/n2 & study design in which the researcher might create &n artificial setting,
control some !ariables and manipulates the independent !ariable to establish cause8and8effect
relationship is called -xperimental Design.
9esson :o. ;
<lassification of Research
&hat is e4)lorator6 research and write down /oals o! this research and also write a-ot
sorces sed -6 e4)lorator6 research !or collection o! in!or'ation"
Soltion
If the issue was new or the researcher has written little on it, you began at the
beginning, this is called 0exploratory research1.
Its may be the first stage in a seuence studies, and useful preliminary step for new
researcher=s. -xploratory research rarely yields definiti!e answers. It addresses the 0what1
uestion i.e. what is this social acti!ity really about, and difficult to conduct because there are
few guidelines to follow.
2oals o! ,4)lorator6 Research
De!elop well grounded picture of the situation.
Determine the feasibility of conducting the study.
%ecome more familiar with the facts, setting, and concerns.
$ormulate uestions and refine issues for more systematic inuiry.
De!elop tentati!e theories, generate new ideas.
,4)lorator6 Research sed !ollowin/ sorces 7 tools !or collectin/ in!or'ation.
-xperience 'ur!eys. In experience sur!ey the researcher tries to contact indi!iduals who
are ,nowledgeable about a particular research problem.
'econdary Data &nalysis. &nother economical and uic, source of bac,ground
information is secondary data analysis. It is preliminary re!iew of data collected for
another purpose to clarify issues in the early stages of a research effort.
<ase 'tudies. "he purpose of case study is to obtain information from one or a few
situations that are similar to the researcher=s problem situation.
#ilot 'tudies. & pilot study implies that some aspect of the research is done on a small
scale.
&hat is .escri)tive Research and write down its /oals and also write its sorces !or
collection o! data.
Soltion
& research that presents a picture of specific details of a situation, social setting, or
relationship is called 0descripti!e research1.
"he ma(or purpose of descripti!e research is to describe characteristics of a population or
phenomenon, and see,s to determine the answers to who, what, when, where, and how
uestion. $or examples labor force sur!eys, population census, and educational census. It
offers to the researcher a profile or description of rele!ant aspects of the phenomenon.
2oals o! .escri)tive Research
<larify seuence, set of stages.
#resent bac,ground information.
4i!e a !erbal or numerical picture of the situation.
<reate a set of categories or classify the information.
$ocus on who, what, when, where, and how, but not why>
.escri)tive Research sed !ollowin/ sorces 7 tools !or collectin/ in!or'ation.
Data 4athering "echniues li,e 'ur!eys.
$ield Research
<ontent &nalysis.
&hat is e4)lanator6 research and write down its /oals"
Soltion.
?hen we encounter an issue that is already ,nown and ha!e a description of it, we
might begin to wonder why things are the way they are. "he desire to ,now 0why1 to explain is
the purpose of explanatory. It builds on exploratory research and descripti!e research and loo,s
for causes and reasons. $or example why parents abuses their children.
-xplain things not (ust reporting such as 0?hy1>
-laborate and enrich a theory=s explanation.
Determine which of se!eral explanations is best.
&d!ance ,nowledge about underlying process.
%uild and elaborate or enrich a theory=s predictions or principle.
T6)es o! A))lied Research
Ma1or t6)es o! a))lied research sed -6 )ractitioners are
Action Research2 "he applied research that treats ,nowledge as a form of power and
abolishes the line between research and social action is called 0action research1. It
incorporates ordinary or popular ,nowledge li,e it focuses on power with a goal of
empowerment see,s to increase awareness and tied directly to political action. &ction
researchers assume that ,nowledge de!elops from experience, particularly the
experience of social political action.
(')act Assess'ent Research2 its purpose is to estimate the li,ely conseuences /
assessment of a planned change and ma,ing choices among alternati!e policies. 'uch
as assessment of %asha dam on the en!ironment.
,valation Research2 it addresses the uestion 0did it wor,1, the process of
establishing !alue (udgment based on e!idence about the achie!ement of the goals of a
program. It measures the effecti!eness of a program or policy and uses se!eral research
techniues li,e sur!ey, field research. "wo types of e!aluation research are
o $ormati!e -!aluation Research2 It is built in monitoring or continuous feedbac,
on a program used for program management.
o 'ummati!e -!aluation Research2 9oo,s at final program outcomes
+o')ared 8asic Research with A))lied Research"
Soltion
@. %asic Research2 "he scientific community is the primary consumer of basic
research.
&pplied Research2 "he consumers of applied research findings are practitioners
such as teachers, counselors, and casewor,ers etc.
A. %asic Research2 basic researchers emphasize high standards and try to conduct
near perfect research.
&pplied Research2 &pplied researcher ma,e more trade offs.
;. %asic Research2 5igh standards applied
&pplied Research2 Buic, and dirty may not meet high standards.
C. %asic Research2 9ogic and rigorous research design
&pplied Research2 &pply to areas of interest of sponsors
D. %asic Research2 'uccess results published impact on other scientists.
&pplied Research2 'uccess results are used by sponsors.
9esson :o. C
"heory and Research
Role o! Theor6$

Theor6 as a 0rientation$ theory plays as a role of orientation because a ma(or

function of a theoretical system is that it narrows the range of facts to be studied.
'uppose any phenomenon or ob(ect may be studied in many different ways.
-ach science and specialization within a broader field abstracts from reality and
broad orientation of each field then focuses upon limited range of things while
ignoring or ma,ing assumptions about others.
Theor6 as a conce)tali3ation and classi!ication 2 each science is organized
by a structure of concepts, which refer to ma(or processes and ob(ects to be
studied. &s a conseuence, a ma(or tas, in any science is the de!elopment of
classification, a structure of concepts, and an increasing precise set of definitions
for these terms.
Theor6 as a s''ari3in/ role2 & further tas, which theory performs is to
summarize concisely what is already ,nown about the ob(ect of study. "hese
summaries may be di!ided into two simple categories 6i7 -mpirical 4eneralization
and 6ii7 systems of relationships between propositions.
Theor6 as a )redicts !acts2 If the theory summarizes facts and states a general
uniformity beyond the immediate obser!ations, it also becomes a prediction of
facts which ha!e se!eral component.
Theor6 as a )oints /a)s in 5nowled/e$ 'ince theory summarizes the ,nown
and predicts facts which ha!e not been obser!ed, it must also point to areas
which ha!e not yet been explored. "heory also points to gaps of a more basic
,ind, while these gaps are being filled, and changes in the conceptual scheme
usually occur.
9ro)ositions2 proposition are statements concerned with the logical
relationships among concepts. & proposition explains the logical lin,age among
certain concepts by asserting a uni!ersal connection between concepts.
Theor6 .evelo)'ent$ is essentially a process of describing phenomena at
increasingly higher le!els of abstraction. "heorist translates their
conceptualization of reality into abstract ideas. "hus theory deals with
obstruction.
&hat is e4)ansion2 the expansion is the whole process through which the
phenomenon emerges, and we would li,e to understand the process to reach
prediction.
Theor6$ "heory is a coherent set of general propositions, used as principles of
explanations of the apparent relationship of certain obser!ed phenomena
9esson :o. D
<oncepts
+once)t
& concept is a generalized idea about a class of ob(ects, attributes, occurrences, or
processes that has been gi!en a name. In other words a concept is an idea expressed as a
symbol or in words.
.e/ree o! A-straction
0"he organization of concepts in seuence from the most concrete and
indi!idual to the most general indicates the degree of abstraction1. .o!ing up the ladder of
abstraction, the basic concept becomes more abstract, wider in scope, and less amenable to
measurement. "he scientific researcher operates at two le!els of concepts and on the empirical
le!el of !ariables. &t the empirical le!el we experience reality and this is we obser!e ob(ects or
e!ents.

(')ortance o! conce)ts
'ocial sciences concepts form a specialized language, or (argon.
'pecialists use (argon as a short hand way to communicate with one another. .ost fields ha!e
their own (argon li,e #hysicians, 9awyers, -ngineers, and .echanics etc. all ha!e specialized
language and use their (argon to refer to the ideas and ob(ects with which they wor,. 'pecial
problems grow out of the need for concept precision and in!enti!eness and !ague meanings
attached to a concept create problems of measurement. "herefore, not only the construction of
concepts is necessary but also these should be precise and the researchers should ha!e some
agreement to its meaning.
Identification of concepts is necessary because we use concepts in hypothesis formulation and
one of the characteristics of a good hypothesis that it should be conceptually clear.
.ictionar6 .e!inition
+once)tal .e!inition
<onceptual definitions are a definition in abstract,
theoretical terms. It refers to other ideas or constructs. It in!ol!es thin,ing carefully, obser!ing
directly, consulting with others, and trying possible definitions. <onceptual definitions are lin,ed
to theoretical framewor,s and to !alue positions. 'ome constructs are highly abstract and
complex and contain lower le!el concepts within them, and which can be made e!en more
specific. /ther concepts are concrete and simple. $or example construct such as age is easier
to define than is a complex abstract concept such as morale.
0)erational .e!inition
In research we must measure concepts which reuire a rigorous definition
and construct a concept must be made operational in order to be measured. &n operational
definition gi!es meaning to 0a concept by specifying the acti!ities or operations necessary to
measure it and specifies what must be done to measure the concept under in!estigation1.
/peration definition also called a wor,ing definition stated in terms of specific testing or
measurement criteria. 'o, in operational definition we must specify concrete indicators that can
be obser!ed / measured.
9esson :o. *
Variables and "ypes of Variables
&hat is varia-le"
Variable is central idea in research. 'imply defined, !ariable is concept that !aries. "here are
two types of concepts, 6i7 those that refer to a fixed phenomenon, and 6ii7 those that !ary in
uantity, intensity, or amount. "he second type of concept and measures of the concept are
!ariables. & !ariable is defined as anything that !aries or changes in !alue. Variables ta,e on
two or more !alue. %ecause !ariable represents a uality that can exhibit differences in !alue,
usually magnitude or strength, and also !ariable generally is anything that may assume different
numerical or categories !alues.
T6)es o! #aria-le
+ontinos #aria-les$ If the !alue of a !ariable can be di!ided into fractions then it is
called 0continuous Variable1. 'uch !ariable can ta,e infinite numbers of !alues. Income,
"emperature, &ge, or a test score are the examples of continuous !ariable. "hese
!ariables may ta,e on !alues within a gi!en range or in some cases an infinite set.
.iscontinos #aria-les$ &ny !ariables that has a limited number of distinct !alues
and which cannot be di!ided into fractions is called 0discontinuous !ariables1. It is also
called categorical, classificatory, and discrete !ariable. 'ome !ariables ha!e only two
!alues, reflecting the presence or absence of a property, li,e employed or unemployed,
male or female ha!e only two !alues and it referred to as dichotomous.
.e)endent #aria-le$ "he cause !ariable or the one that identifies forces or conditions
that act on something else is called 0Dependent Variable1. Dependent !ariables
0depends on1 the causes and also referred to as 0<riterion Variable1. In statistical
analysis dependent !ariable identified by the symbol 6E7. "he stoc, mar,et price of the
product is the example of dependent !ariable.
(nde)endent #aria-les$ "he !ariable that is the effect or the result or outcome of
another !ariable is called 0independent !ariable1. It is also called outcome or effect
!ariable. "he independent !ariable is 0independent of1 prior causes that act on it. In
statistical analysis independent !ariable identified by the symbol 6F7. "he success of the
new product is the example of independent !ariable.
Moderatin/ #aria-les$ & !ariable that has a strong contingent effect on the dependent
and independent !ariables relationship is called 0moderating !ariables1. "hat is, the
presence of ;
rd
!ariable modifies original relationship between the dependent and
independent !ariable. $or example a strong relationship has been obser!ed between the
uality of a library facility 6F7 and performance of the student 6E7.
(ntervenin/ #aria-les
& basic casual relationship reuires only independent and dependent !ariable. & third
type of !ariable, 0the inter!ening !ariable1, appears in more complex casual relationship.
It comes between independent and dependent !ariables and shows the lin, or
mechanism between them. In sense, the inter!ening !ariable acts as a dependent
!ariable with respect to independent !ariable and acts as an independent !ariable
toward the dependent !ariable.
,4traneos #aria-les
&n almost infinite number of extraneous !ariables exist that might concei!ably affect a
gi!en relationship. 'uch !ariables ha!e to be identified by the researcher. In order to
indentify the true relationship between the independent and dependent !ariables, the
effect of the extraneous !ariables may ha!e to be controlled. <onfounding factors is
another name used for extraneous !ariable.
9esson :o. +
5ypothesis "esting and <haracteristics
:6)othesis
& hypothesis may be defined as 0a logically speculation relationship between two or
more !ariables, expressed in the form of a testable statement is called 05ypothesis1.
Relationship is proposed by using a strong logical argumentation and may be part of theoretical
framewor, of the study.
T6)es o! :6)othesis
.escri)tive :6)othesis2 Descripti!e hypothesis contains only one !ariable thereby it is
called as uni!ariate hypothesis. It is typically state the existence, size, form, or
distribution of some !ariable. *or e4a')le what is the le!el of (ob commitment of the
officers in my organization>
Relational :6)othesis$ these are the propositions that describe a relationship between
two !ariables. "he relationship could be directional or non directional, positi!e or
negati!e, and causal or simply <orrelational. The directional h6)othesis is the one in
which the direction of the relationship has been specified. The non directional
h6)othesis is the one in which the direction of the relationship has not been specified.
+orrelational :6)othesis simply state that the !ariables occur together in some
specified manner without implying that one causes the other. *or e4a')le le!el of
commitment of the officers is positi!ely associated with their le!el of efficiency.
,4)lanator6 7 +ases :6)othesis implies the existence of or a change in one !ariable
causes or leads to a change in the other !ariable. "his brings in the notions of
independent and the dependent !ariables. <ause means to 0help ma,e happen1. 'o, the
independent !ariable may not be the sole reason for the existence of or change in the
dependent !ariable, therefore researchers may ha!e to identify other possible causes
and control their effect in case the casual effect of independent !ariable has to be
determined on the dependent !ariable and possible in an experimental design of
research.
Nll :6)othesis$ It is used for testing the hypothesis formulated researcher. "he null
hypotheses simply states that there is no relationship between the !ariables or the
relationship between the !ariables is 0zero1. "hat is how symbolically null hypothesis is
denoted as 05G1.
*or e4a')le
:0 ; there is no relationship between the le!el of (ob commitment and the le!el of
efficiency. /r
:o; the relationship between the le!el of (ob commitment and the le!el of efficiency is
zero.
"he two !ariables are independent of each other.
-xperts recommended that we test out hypothesis indirectly by testing the null
hypothesis. "he re(ection of the null hypothesis leads to the acceptance of the
alternati!e hypothesis.
Alternative :6)othesis2 "he alternati!e hypothesis simply states that there is a
relationship between the !ariables under study. &lternati!e hypothesis symbolically
denoted as 05@1. *or e4a')le
5@H there is a relationship between the le!el of (ob commitment of the officers and their
le!el of efficiency.
Research :6)othesis$ Research hypothesis is the actual hypothesis formulated by the
researcher which may also suggest the nature of relationship.
9esson :o. I
Re!iew of 9iterature
2oals o! a literatre Review
To de'onstrate a !a'iliarit6 with a -od6 o! 5nowled/e and esta-lish
credi-ilit62 & good re!iew increases a reader=s confidence in the researcher=s
professional competency, ability, and bac,ground, also tells that the researcher
,nows the research in an area and the ma(or issues.
To 5now the )ath o! )rior research and how a crrent research )ro1ect is
lin5ed to it2 & good re!iew places a research pro(ect in a context and
demonstrates its importance by ma,ing connections to a body of ,nowledge and
also indicates the direction, ability, and bac,ground of research.
To inte/rate and s''ari3e what is 5nown in an area$ & good re!iew points
out areas where prior studies agree, where they disagree, where ma(or uestions
remain and pulls together and produces different results, and also indicates the
direction for future research. :o wastage of effort.
To learn !ro' others and sti'late new ideas$ & good re!iew identifies blind
alleys and suggest hypotheses for replication and tells procedures, techniues,
and research design. &lso tells what others ha!e found so that a researcher can
benefit from the others efforts.
(denti!ication o! #aria-les 2 & good re!iew identified different important
!ariables that are li,ely to influence the problem situation are not left out of the
study.
:el)s in develo)in/ theoretical !ra'ewor5$ & good re!iew may be pro!ide a
ready made theoretical model of research.
T6)es o! Reviews
Sel! Std6 Review$ It increases the reader=s confidence and demonstrates
familiarity with an area is rarely published but it often is part of an educational
program. It gi!ing to others confidence in a re!iewers command of field, and is a
side benefit that building the re!iewers self confidence.
+onte4t Review$ It places a specific pro(ect in the big picture and one of the
goals of re!iew is creating a lin, to a de!eloping body of ,nowledge. "his is a
bac,ground or context re!iew. It introduces the rest of a research and establishes
the significance of a research uestion and also tells how a pro(ect fits into the
big picture.
:istorical Review$ It traces the de!elopment of an idea or shows how a
particular issue or theory has e!ol!ed o!er time. Researchers conduct historical
re!iew only on the most important ideas in a filed.
Theoretical Review$ It presents different theories that assert to explain the
same thing, then how well each accounts for findings. Researcher=s also use it to
integrate two theories. It sometimes forms a hybrid or the historical theoretical
re!iew.
Methodolo/ical Review$ In it researcher=s e!aluates the methodological
strength of past studies. It describes conflicting results and shows how different
research designs, samples, measures and so on account for different results.
9esson :o. @I
<riteria for 4ood .easurement
Validity2 Validity is the ability of an instrument to measure what it is supposed to
measure.
<ontent Validity2 "he content !alidity of a measuring instrument is the extent to
which it pro!ides adeuate co!erage of the in!estigati!e uestions guiding the
study. If the instrument contains a representati!e sample of the uni!erse of
sub(ect matter of interest, then the content !alidity is good.
9esson :o. @J
Research Design
Research .esi/n
& research design is a master plan specifying the methods and procedures for
collection and analyzing the data. It is a strategy or blueprint that plans the action
for carrying through the research pro(ect data and It also in!ol!es a series of
rational decision ma,ing choices depending upon the !arious options a!ailable to
the researcher=s. %roadly it is composed of different elements li,e,
.i!!erent ,le'ents o! Research .esi/n
9r)ose o! Std6$ $rom the perspecti!e of purpose of the study, a research
can be exploratory, descripti!e and explanatory. %eyond the e4)lorator6 sta/e
now we are entering into the formal stage of delineate the plan for data
collection, data processing, and data analysis. If the research is concerned with
finding out who, what, where, when, or how much, then the study is descri)tive.
If it is concerned with learning why, that is how one !ariable produces changes in
another it is casal. If we try to explain relationship among !ariables then the
study is e4)lanator6.
<nit o! Anal6sis$ "he unit of analysis refers to the le!el of aggregation of the
data collected during the subseuent data analysis stage. )nits of analysis in a
study are typically also the units of obser!ation. If the researcher is interested in
studying two person interactions, then se!eral two person groups ,nown as
dyads will become the unit of analysis. If the problem statement is related to
group effecti!eness, the unit of analysis would be at /ro) level. If we compare
different departments in the organization, then data analysis will be done at the
de)art'ent level.
Ti'e .i'ension$ ?hen we ma,e the obser!ations more or less at one time or
o!er a long period, former called as +ross Sectional stdies and the latter as
lon/itdinal stdies. +ross Sectional Stdies2 is carried out once and
represents a snapshot of one point in time and data are collected (ust once, o!er
a period of days or wee,s or months in order to answer the research uestion.
Lon/itdinal Stdies2 are representing o!er an extended period and it can
trac, changes o!er time. It can be panel studies and cohort studies.
Research control o! #aria-les$ In terms of researcher=s ability to manipulate
!ariables, we can differentiate between e4)eri'ental and e4 !acto desi/n.
,4)eri'ental .esi/n2 In an experiment the researcher attempts to control and /
or manipulate the !ariables in the study. -xperimental design is appropriate
when one wish to disco!er certain !ariables produce effects in other !ariables
and pro!ides the most powerful support possible for hypothesis of causation. It
can be contri!ed and non contri!ed. ,4 *acto .esi/n2 In!estigators ha!e no
control o!er the !ariables in the sense of being able to manipulate them> "hey
can only report what has happened or what is happening. 'ur!ey research is an
example of such study.
+hoice o! Research .esi/n2 "hey could be number of ways to collect the data
depending upon whether the study is uantitati!e or ualitati!e, descripti!e or
explanatory, cross sectional or longitudinal, and contri!ed or non contri!ed, the
researcher decides about the mode of obser!ation.
Sa')lin/ .esi/n2 "he basic idea of sampling is that by selecting some of the
elements in population, we may draw conclusions about the entire population.
'ampling has its own ad!antages and disad!antages, and depending upon the
nature of the study the researcher=s decides about sampling design.
*ield .ata +ollection$ Depending upon the mode of obser!ation, the
researcher will outline the procedure for field operations.
.ata 9rocessin/ and .ata Anal6sis$ In the research design the researcher is
reuire telling how the data shall be processed and analysis plans explicated. In
case the ualitati!e data are to be uantifies the procedures should be spelled
out.
Ste)s in condctin/ Srve6 Research
.evelo) the Srve6 (nstr'ent2 "he researcher=s de!elops an instrument li,e a
sur!ey uestionnaire or inter!iew schedule that researcher=s uses to measure !ariables.
Respondents read the uestions themsel!es and mar, answers on a uestionnaire or a
set of uestions read to the respondent by an inter!iewer, who also records the
responses. "o simplify the discussion, we will use only the term uestionnaires.
9lan how to record data$ ?hen preparing the uestionnaire, the researcher thin,s
ahead to how will record and organize data for analysis.
.ecide on tar/et )o)lation= /et sa')lin/ !ra'e= decide on sa')le si3e= and
select the sa')le.
Locate Res)ondents$ "he researcher locates sampled respondents in person, by
telephone, or by mail who gi!en information and instructions on completing the
uestionnaire or inter!iew.
.escri-e 'ethods and !indin/s in research re)ort= )resent !indin/s to others !or
anal6sis and evalation.
9esson :o. AG
'ur!ey Research
9ersonal (nterview$ & personal inter!iew is a two way con!ersation initiated by an
inter!iewer to obtain information from a respondent and differences in the role of the
inter!iewer and the respondent are pronounced. #ersonal may ta,e place in a factory, in
a homeowner=s doorway, in an executi!e=s office, in a shopping mall or in other settings.
Advanta/es o! 9ersonal (nterview
The o))ortnities !or !eed-ac5$ #ersonal inter!iews allow for feedbac,. "he
inter!iewer pro!ides feedbac, in clarifying any uestions an employee. &t the
conclusion of the inter!iew, respondent gi!en additional information concerning
the purpose of the study.
9ro-in/ +o')le4 Qestion$ &n important characteristic of personal inter!iew
is the opportunity to follow up by probing. #robing becomes the more important
when the uestions don=t ha!e structured response categories. "he complex
uestion that cannot easily be as,ed in telephone or mail sur!eys can be
handled by s,illful inter!iewers.
Len/th o! (nterview$ If the researcher ob(ecti!e reuires an extremely lengthy
uestionnaire, personal inter!iews may be the only alternati!e.
9ro)osed and #isal Aids$ Inter!iewing respondents face to face allows an
in!estigator to show them a new product sample, a s,etch of proposed office, or
some other !isual aid. "he respondents can e!en taste samples of different
products and can gi!e their e!aluations.
:i/h 9artici)ation Rate$ In personal inter!iews there is a higher rate of
participation rate of the respondents compared with mail sur!eys and telephone
inter!iews. ?hile some people are reluctant to participate in a sur!ey, the
presence of an inter!iewer generally increases the percentages of people willing
to complete the inter!iew. .ost people en(oy sharing information and insights
with friendly and sympathetic inter!iewers.
:on 9iterates can #articipate in 'tudy2 :ice the respondent has neither to read
nor to write, therefore, an illiterate person can also ta,e part in the sur!ey study.
+o')ter Assistant 9ersonal (nterviewin/2 ?ith the use of modern
technology the responses of the respondent can be entered into a portable
microcomputer to reduce error and cost.
Disad!antages of #ersonal Inter!iew
:i/h +ost2 #ersonal inter!iews are generally more expensi!e than mail,
internet, and telephone sur!eys because the training of the field inter!iewers,
super!ision, and other logistical support cost may add up the total cost of the
study. #eople usually estimate the cost of personal inter!iews is usually @D times
higher than the mail sur!ey.
Lac5 o! Secrec6 o! Res)ondent2 %ecause the respondent in a personal
inter!iew is not un,nown therefore he or she may be reluctant to pro!ide
confidential information to another person.
+all-ac5s$ ?hen the person selected to be in the sample cannot be contacted
on the first !isit, a systematic procedure is normally initiated to callbac, at
another time which is a labor intensi!e wor, and definitely increases the cost.
(nterviewer (n!lence2 "here is some e!idence that the demographic
characteristics of the inter!iewer influence respondents= answers.
No 0))ortnit6 to +onslt$ "he inter!iew may ta,e place anywhere, place of
wor,, in the shopping mall, and at home the respondent may be unable to
consult record, in case he or she has to do so for any specific uestion.
So'e Nei/h-orhoods are .i!!iclt to #isit$ Due to security reasons some
neighborhoods may not allow outsiders to enter the permission.
9esson :o. A@
Intercept Inter!iews in .all and other 5igh "raffic &reas
Tele)hone (nterviewin/ $ "elephone inter!iewing has been a main stay of commercial
sur!ey research and the uality of data obtained by telephone may be comparable to
that collected in personal inter!iew. "elephone sur!ey pro!ides representati!e samples
of general population in most industrialized countries.
Stren/ths o! Tele)hone (nterviewin/
:i/h S)eed2 "he speed of data collection is a ma(or ad!antage of telephone
inter!iewing, whereas data collection with mail or personal inter!iew can ta,e se!eral
wee,s. ?hen the inter!iewer enters the respondents= answers directly into a computer
system, data processing can be done e!en faster.
Save +ost2 &s the cost of personal inter!iew continues to increase, telephone
inter!iews are becoming relati!ely inexpensi!e, estimated ADK less than personal
inter!iew.
+all-ac5s$ &n unanswered call a respondent who is not at home reuires a callbac,.
"elephone callbac,s are substantially easier and less expensi!e than personal inter!iew
callbac,s.
8etter Access too hard to Reach Res)ondents2 'ome people don=t want to go for
person to person inter!iewing and may be reluctant to !isit certain neighborhoods,
especially in the e!ening, so, telephone inter!iewing o!ercomes such problems.
+o')ter Assisted Tele)hone (nterviewin/$ Responses can be directly entered into
computer file to reduce error and cost.
&ea5nesses o! Tele)hone (nterviewin/
A-sences o! *ace to *ace +ontact$ "elephone inter!iews are more impersonal than
face to face inter!iews. &bsence of face to face contact can be a liability. "he inter!iewer
and the respondents don=t see each other what they are doing.
Res)onse Rate is lower than )ersonal interviews$ 'ome indi!iduals refuse to
participate in telephone inter!iews. 'o, response rate in telephone inter!iews is lower
than personal inter!iews.
Lac5 o! #isal Medi'$ Researcher reuiring !isual material cannot be conducted by
phone.
Li'ited .ration$ 9ength of inter!iew limited. Respondents may hang up when they
feel spent up.
.istractin/ 9h6sical ,nviron'ent$ .ultiple phones distract the inter!iew situation
which may affect the uality of the data.
Mali Qestionnaire2 & mail sur!ey is a self administered uestionnaire sent to respondents
through the mail. "his paper and pencil method has se!eral ad!antages and disad!antages.
Advanta/es o! Mail Qestionnaire
2eo/ra)hic *le4i-ilit62 .ail uestionnaires can reach a geographically dispersed
sample separately and at a reasonably low cost because inter!iewers are not reuired.
Respondent=s those who are difficult to reach can be contacted more easily by mail.
Sa')le Accessi-ilit6$ Researchers can contact participants who may otherwise be
inaccessible. %ut researchers can often access these special participants by mail or
computer.
Sel! Ad'inistrated Qestionnaires
Save +ost$ .ail uestionnaires are relati!ely economical compared to personal
inter!iews and telephone sur!eys. 5owe!er, these may not be so cheap.
Standard Qestions$ .ail uestionnaires are highly standardized, and the uestions
are uite structured.
.isadvanta/es o! Mail Qestionnaires
Low Res)onse Rate2 .ail Buestionnaire has !ery low rate of return of the filled
uestionnaires.
Low +o')letion Rate2 "here are chances that respondents lea!e many uestions as
unanswered, either because they did not understand the uestion.
(ncreases +ost$ "he researcher ,eeps on waiting for the return. ?hen enough
response is not there, then the reminders are sent. ?ith the reminders copies of the
uestionnaires are sent then all this adds to the cost of the study.
9esson :o. AA
'elf &dministered Buestionnaires
+over Letter2 "he co!er letter that con!oys the uestionnaire or is printed on the first
page of the uestionnaires is an important means of inducing a reader to complete and
return the uestionnaire. & personalized co!er letter addressed to a specific indi!idual
show the respondent that he or she is important and indi!idually typed letter on letter
head !ersus printed form is an important element in increasing the response rate in mail
sur!eys.
$ollow <)$ *ollow up implies the communication of the message to respondents
through different means of the return of uestionnaire. & follow up may include a
duplicate uestionnaire or may merely reminder to return the original uestionnaire.
.ultiple contacts almost always increase response rates.
Mone6 :el)s2 "he respondent=s moti!ation for returning a uestionnaire may be
increased by offering monetary incenti!es or premiums, such as pens, lottery tic,ets,
and !ariety of premiums. .oney incenti!e wor,s for all income categories.
Srve6 S)onsorshi)2 'ponsorship of the study ma,es a difference for moti!ation the
respondents to return the uestionnaires. It depends up the goodwill of the sponsoring
agency that can acti!ate or deacti!ate the respondent to fill the uestionnaire and return
it. 'ponsorship by well ,nown and prestigious organization such as uni!ersities or
go!ernment agencies may significantly influence response rates.
(nterestin/ Qestions2 In the topic of research certain interesting uestion can be
added to the uestionnaire, perhaps in the beginning to stimulate the respondent=s
interest and to induce cooperation.
8ene!its o! ,'ail Srve6s2 include speed of distribution, lower distribution, processing
cost, faster turnaround time, more flexibility, and less handling of paper uestionnaires.

(nternet Srve6s2 &n internet sur!ey is a self administrated uestionnaires posted on a

web site. Respondents pro!ide answers to uestion displayed on screen by highlighting
a phrase, clic,ing an icon, or ,eying in an answer.
Advanta/es o! (nternet Srve6s
S)eed and +ost ,!!ective2 Internet sur!ey allow the mar,eters to reach a large
audience, to personalize the indi!idual messages, and to secure confidential
answers uic,ly and cost effecti!e because computer to computer self administrated
uestionnaires eliminated the cost of paper, postage, data entry, and other
administrati!e costs.
#isal A))eal and (nteractivit62 Internet sur!eys can be interacti!e because the
researcher can use more sophisticated lines of uestioning based on the
respondents prior answers and excellent medium fort li,e presentation of !isual
materials, such as photographs, ad!ertisements, and mo!ie trailers.
+all-ac5s$ ?hen the sample for internet sur!ey is drawn from a consumer panel, it
is easy to recontact those who ha!e not yet completed the uestionnaire.
Accrate Real Ti'e .ata +a)tre$ &ccurate real time data capture allows for real
time data analysis. & researcher can re!iew up to the minute sample size counts and
tabulation data from an internet sur!ey in real time.
Res)onse Rate$ Response rate can be increased by sending email friendly
reminders.
.isadvanta/es o! (nternet Srve6s
All 9eo)le +annot 9artici)ate2 .any people in the general public cannot access to
internet and all people with internet access don=t ha!e the same le!el of technology.
.any lac, powerful computer or software that is compatible with ad!anced features
programmed into many internet uestionnaires. 'ome indi!iduals ha!e minimum
computer s,ills and may not ,now how to na!igate through and pro!ide answers to
internet uestionnaires.
No 9h6sical (ncentive2 )nli,e mail or internet sur!eys don=t offer the opportunity to
send a physical incenti!e to the respondent.
9esson :o. A;
"ools for Data <ollection
Tools *or .ata +ollection2 %roadly there are three tools for data collection as part of
communication sur!eys. "hese are
(nterview Schedle2 Inter!iew schedule is predesigned list of uestion used for
communication with the respondent. In case of inter!iew schedule, the list of uestions
remains in the hands of the inter!iewer, he as,s uestions from the respondent, and
gets his or her responses and recorded it.
Qestionnaire$ Buestionnaire is also a predesigned list of uestions used for
communication with the respondent. Buestionnaire is also a list of uestions, which
handed o!er to the respondent, who reads the uestions and records the answers
himself.
(nterview 2ide$ It is a list of topics that are to be co!ered during the course of
inter!iew. Inter!iew guide is used for purposes for an in depth inter!iewing. In inter!iew
guide uestions on the topics are formulated on the spot and most of the uestions are
open ended.
2idelines !or Qestionnaire .esi/n2 Buestionnaire design is one of the most critical stages
in the sur!ey research process. ?hile common sense and good grammar is reuired in the art
of uestionnaire design. $urther proper wording the uestionnaire is crucial, as some problems
may be minimized or a!oided altogether if a s,illed researcher composes the uestions. & good
uestionnaire forms an integrated whole.
Qestionnaire Relevanc62 & uestionnaire is rele!ant if the information that is needed
to sol!e the problem is obtained. &s,ing the wrong or irrele!ant uestion is a pitfall to be
a!oided. If the tas, is to pinpoint compensation.
Qestionnaire Accrac6$ &ccuracy means that the information is reliable and !alid.
/btaining accurate information from respondents is strongly influenced by the
researcher=s ability to design a uestionnaire that facilities recall and that will moti!ate
the respondent to cooperate. "herefore a!oid (argon, slang, and abbre!iations and
words used in the uestionnaire should be readily understandable to all respondent.
Avoid .o-le 8arreled Qestion$ .a,e each uestion about one and only one. &
double barreled uestion consists of two or more uestions (oined together and ma,es
the respondent=s answer unclear.
Avoid Leadin/ Qestions$ .a,e respondent=s feel that all responses are legitimate.
Don=t le them aware of an answer that the researcher wants. & leading uestion is the
one that leads the respondent=s to choose one response o!er another by its wording.
<se *ilter Qestion$ $ilter Buestion is that uestion which screens out respondent=s
not ualified to answer a second uestion. $ilter uestions minimizes the chances of
as,ing uestions that are inapplicable.
La6ot o! the Qestionnaire2 4ood layout and physical attracti!eness is crucial in
mail, internet, and other self administrated uestionnaires. "here are two format or
layout issues 6i7 the o!erall physical layout of the uestionnaire and 6ii7 the format of
uestions and responses.